Alcohol use disorders (AUDs) have a massive impact on public health, and are associated with risk for many additional disorders. A host of biological (genetic) and environmental factors interact throughout the addictive process. As a result, despite being highly heritable, the genetic determinants of risk remain largely unknown, hindering advances in AUD prevention and treatment. An important goal of this proposal is to progress toward novel therapeutic approaches using robust preclinical (mouse) models. Unique advantages of this proposal include that our data were the first to implicate Mpdz (encoding the multiple PDZ protein MUPP1) in any behavior, to prove that it significantly affects translational alcohol behavior (risk for physiological dependence withdrawal), and also to provide evidence implicating Mpdz in alcohol consumption. We propose: (1) To test whether Mpdz affects clusters (or not) of translational behaviors. We will complete testing our novel Mpdz+/- knockout heterozygote and Mpdztg transgenic genetic models for translational alcohol behaviors (i.e., two- bottle choice preference drinking, binge drinking, chronic intermittent drinking; conditioned place preference (CPP); withdrawal convulsions and anxiety) and important primary translational and control behaviors (water, quinine, and saccharin consumption; control CPP; baseline convulsions, anxiety, and fear conditioned learning and memory) to be more informative about the nature of Mpdz actions. Viral methods will be used to dissociate direct Mpdz actions in the adult brain from its potential actions during development, and provide additional mechanistic (e.g., brain regional) information. (2) To increase mechanistic understanding toward new therapeutic approaches. We propose an approach to simultaneously and thus more rapidly increase Mpdz mechanistic understanding and, importantly, detect new mechanistic targets with therapeutic potential. Utilizing our novel genetic models and wild-type littermates, and using appropriate alcohol and control groups, we will begin using unbiased whole genome coexpression network analysis to elucidate coexpression patterns (indicative of functional networks, including upstream regulatory and downstream compensatory mechanisms, providing additional mechanistic information) that are alcohol responsive, Mpdz responsive, both, or neither. The integration of our results with other data (e.g., using HitWalker an innovative algorithm created by our team for the integration of data from a wide variety of sources, including known drug-protein/gene interactions), represents an innovative and tremendously powerful approach to simultaneously elucidate mechanistic bases of translational alcohol behaviors and key genetic influences to point to new therapeutic approaches. (3) Rigorously test the most promising new therapeutic approaches using robust preclinical (mouse) models. Importantly, we have the expertise to take full advantage of emerging small molecule and pharmacological approaches and genetic tools as needed, and to assess their efficacies to affect translational behaviors. This represents a crucial step in the testing of promising new therapies for the human clinical condition.